Small Batch Production Research Articles

A Review of Automotive Spare-Part Reconstruction Based on Additive Manufacturing

In the Industry 4.0 scenario, additive manufacturing (AM) technologies play a fundamental role in the automotive field, even in more traditional sectors such as the restoration of vintage cars. Car manufacturers and restorers benefit from a digital production workflow to reproduce spare parts that are no longer available on the market, starting with original components, even if they are damaged. This review focuses on this market niche that, due to its growing importance in terms of applications and related industries, can be a significant demonstrator of future trends in the automotive supply chain. Through selected case studies and industrial applications, this study analyses the implications of AM from multiple perspectives. Firstly, various types of AM processes are used, although some are predominant due to their cost-effectiveness and, therefore, their better accessibility and wide diffusion. In some applications, AM is used as an intermediate process to develop production equipment (so-called rapid tooling), with further implications in the digitalisation of conventional primary technologies and the entire production process. Secondly, the additive process allows for on-demand, one-off, or small-batch production. Finally, the ever-growing variety of spare parts introduces new problems and challenges, generating constant opportunities to improve the finish and performance of parts, as well as the types of processes and materials, sometimes directly involving AM solution providers.

A Heuristic Integrated Scheduling Algorithm via Processing Characteristics of Various Machines

Complex products with a tree-like structure usually have an asymmetric structure. Therefore, in order to avoid the separation of equipment and operation correlation during the scheduling, the structural attributes of products and the use of equipment resources should be fully considered. However, this feature is ignored in the current research works on the scheduling of multi-variety and small batch products. This leads to increased idle time for equipment and an extended makespan for products. To avoid this situation, a heuristic integrated scheduling algorithm via processing characteristics of various machines (HIS-PCVM) is proposed. In the integrated scheduling, the algorithm first divides the equipment into two categories: the special equipment and the general equipment according to the resources of the production scheduling system. Then, different scheduling methods are designed according to the equipment categories. The makespan of the product is further optimized through various combination methods. Moreover, the constraint audit strategy to guarantee the constraint relationship between the operations is optimized. The earliest scheduling time strategy is proposed to improve the parallelism and serial tightness of the operations. These strategies reduce the idle time of equipment effectively. Experimental results show that the proposed algorithm has a better application effect in reducing the makespan of complex products, both with asymmetric structures and symmetric structures. This also shows that the algorithm is effective in improving the utilization rate of equipment.

Batch sizing control of a flow shop based on the entropy-function theorems

It is a crucial decision-making issue to determine the batch size in production for manufacturing enterprises which implement multi-variety and small batch production. Regarding the current limitations, little research has been done on the relationship between batch sizing and the amount of information needed for production management. This paper first constructs the functional expression between the number of batches and the system entropy according to the measured property of entropy itself. Then, the two theorems on the relationship between the number of batches and the system entropy are theoretically put forward for the first time, i.e., an increase in the number of batches will reduce the system entropy, thus reducing the information required to manage production. Also, the two theorems are fully proved based on the proposed functions. In an empirical study, the variations between the system entropy functions and the number of batches are comparatively analyzed by designating two cases of 10 and 20 machining units. Consequently, the results of graphical representation once again illustrate the validity of the developed two theorems. The conclusions of this study not only provide a crucial information-theoretic basis for the feasible arrangement of batch size with an expert decision support system but contribute to guiding practical production for assembly line managers.

A contrastive learning‐based attention generative adversarial network for defect detection in colour‐patterned fabric

AbstractThe pattern style of colour‐patterned fabrics is varied. Defective fabric samples are scarce in the production of small batches of colour‐patterned fabrics. Therefore, the unsupervised defect‐detection method for colour‐patterned fabric has attracted wide attention. Several unsupervised defect‐detection methods for colour‐patterned fabrics based on convolutional neural networks have been proposed. However, convolutional neural network methods cannot learn long‐range semantic information interaction well because of the intrinsic locality of convolution operations. Besides, as the number of layers in the convolutional neural network increases, the feature maps become more and more complex. Convolutional neural networks experience difficulties in coordinating numerous parameters and extracting key features from complex feature maps. Both these problems reduce the accuracy of the model for detecting defects in colour‐patterned fabrics. In this paper, we propose a Contrastive Learning‐based Attention Generative Adversarial Network (CLAGAN) for defect detection in colour‐patterned fabrics. The CLAGAN possesses two important parts: contrastive learning and a channel attention module. Contrastive learning captures long‐range dependencies by calculating the cosine similarity between different features. The channel attention module assigns different weights to each channel of the feature maps, and it enables the model to extract key features from those feature maps. The experimental results verified the effectiveness of the CLAGAN. It obtained values of 38.25% for intersection over union and of 51.67% for the F1‐measure on the YDFID‐2 public dataset.

A multi-dimensional cognitive framework for cognitive manufacturing based on OAR model

With the production system shifting to a multi-variety and small-batch production mode, the production process faces more user requirements, changes, and uncertainties. To solve the above problems, it is necessary to obtain the status and trend changes information and provide information support for the optimization of decision-making and dynamic adjustment of the production system. However, the production system cognition faces the problems of state coupling, state dynamic transfer and transition, and multi-system interweaving, which makes the production system cognition face huge challenges. Combining technologies such as the Internet of Things, industrial big data, and artificial intelligence, cognitive manufacturing can realize dynamic cognition of the production process, support dynamic adjustment, and become a promising way to solve the dynamic changes and uncertainties of production systems. In addition, as a formal expression of information processing and knowledge learning process in cognitive informatics, the Object-Attribute-Relation (OAR) model can effectively guide the construction of the production process cognitive mechanism. Therefore, this paper proposes a multi-dimensional cognitive framework based on OAR model of the human cognitive world for the dynamic cognitive needs of production system. The framework carries out dynamic cognition from the three dimensions of the manufacturing unit, production situation, and production system, and builds the continuous cognitive abilities from the three dimensions of analysis, decision-making, and learning. By integrating intelligent algorithms in the fields of artificial intelligence, a computable digital twin model is constructed as a carrier to provide the cognitive enabling technologies and capabilities for the production system. Finally, the feasibility of the proposed framework is illustrated by the developed computational digital twin platform. The computable digital twin platform provides the production system with important cognitive capabilities such as states perception, trend prediction, optimization decision-making, and knowledge learning, to support the dynamic cognition and optimization decision-making of the production system, and lay a technical foundation for adaptive production and cognitive manufacturing.

Management Decisions in Multi-Variety Small-Batch Product Manufacturing Process

Management Decisions in Multi-Variety Small-Batch Product Manufacturing Process

Dynamic Performance Prediction in Batch-Based Assembly System with Bernoulli Machines and Changeovers

Worldwide competition and diverse demand of customers pose great challenges to manufacturing enterprises. How to organize production to achieve high productivity and low cost becomes their primary task. In the mean time, the rapid pace of technology innovation has contributed to the development of new types of flexible automation. Hence, increasing manufacturing enterprises convert to multi-product and small-batch production, a manufacturing strategy that brings increased output, reduced costs, and quick response to the market. A distinctive feature of small-batch production is that the system operates mainly in the transient states. Transient states may have a significant impact on manufacturing systems. It is therefore necessary to estimate the dynamic performance of systems. As the assembly system is a typical class of production systems, in this paper, we focus on the problem of dynamic performance prediction of the assembly systems that produce small batches of different types of products. And the system is assumed to be characterized with Bernoulli reliability machines, finite buffers, and changeovers. A mathematical model based on Markovian analysis is first derived and then, the analytical formulas for performance evaluation of three-machine assembly systems are given. Moreover, a novel approach based on decomposition and aggregation is proposed to predict dynamic performance of large-scale assembly systems that consist of multiple component lines and additional processing machines located downstream of the assemble machine. The proposed approach is validated to be highly accurate and computationally efficient when compared to Monte Carlo simulation.

An Improved Model of Product Classification Feature Extraction and Recognition Based on Intelligent Image Recognition.

With the development of the new generation of technological revolution, the manufacturing industry has entered the era of intelligent manufacturing, and people have higher and higher requirements for the technology, industry, and application of product manufacturing. At present, some factories have introduced intelligent image recognition technology into the production process in order to meet the needs of customers' personalized customization. However, the current image recognition technology has limited capabilities. When faced with many special customized products or complex types of small batch products in the market, it is still impossible to perfectly analyze the product requirements and put them into production. Therefore, this paper conducts in-depth research on the improved model of product classification feature extraction and recognition based on intelligent image recognition: 3D modeling of the target product is carried out, and various data of the model are analyzed and recorded to facilitate subsequent work. Use the tools and the established 3D model tosimulate the parameters of the product in the real scene, and record them. Atthe same time, various methods such as image detection and edge analysis areused to maximize the accuracy of the obtained parameters, and variousalgorithms are used for cross-validation to obtain the correct rate of the obtaineddata, and the standard is 90% and above. Build a data platform, compare simulated data with display data by software and algorithm, and check by cloud computing force, so that the model data can be as close to the parameters of the real product as possible. Experimental results show that the algorithm has high accuracy and can meet the requirements of different classification prospects in actual production.

Lifecycle maintenance scheduling of station with multiple categories of information

This paper studies the lifecycle maintenance scheduling of the station under variable and uncertain working conditions. In the multi-specification and small-batch production, the variable production plan leads to the piecewise constant working conditions, and then the piecewise failure process of the station. A general model based on the modality composition method is developed to generate the piecewise failure rate function of the station. To present the impacts of the working conditions on the failure process, the condition-modality theory is introduced. Based on that, an experiment for the above general model is given, and the modality fusion method for the expected failure rate modality is developed. In the maintenance scheduling with uncertain working conditions, the sharing principle is introduced to incorporate the lifecycle information into the dynamic maintenance scheduling. It is observed in the comparisons that the dynamic maintenance scheduling with the sharing principle is more cost-effective than the final payment principle and the expected maintenance scheduling. The discussion suggests that the complete equal instalment principle, which is a simplified version of the sharing principle, is applicable in the industry.

Research of Flexible Assembly Job-Shop Batch-Scheduling Problem Based on Improved Artificial Bee Colony.

This study examined the flexible assembly job-shop scheduling problem with lot streaming (FAJSP-LS), common in multivariety and small-batch production, such as household electrical appliances. In FAJSP-LS, an assembly stage is appended to the flexible job shop, and jobs in the first stage are processed in a large batch to reduce switching costs, while leading to more waiting time, especially during the assembly stage. This article considered splitting the batch into a few sub-batches of unequal and consistent sizes to allow jobs to efficiently pass the two-stage system. With this objective, the problem was modeled as a mixed-integer linear program comprising the following two subproblems: batch splitting and batch scheduling. As the integrated problem is NP-hard, the improved bioinspired algorithm based on an artificial bee colony was proposed, including a four-layer chromosome–encoding structure to describe the solution, as well as an optimization strategy utilizing different bee colonies to synchronously solve this two-stage problem. To examine the algorithm’s efficiency, a benchmark case was used to show that better solutions can be acquired with the improved algorithm regardless of whether the batch was split into equal or unequal sizes. To promote practical implementation, the algorithm was applied to a real case refrigerator workshop and showed better performance on time efficiency when jobs were split into unequal sizes compared to jobs without splitting or splitting into equal sizes.

"A Review for the Formability of Single-Point Incremental Forming (SPIF) Process "

Traditional manufacturing operations require a long time and capital for prototype development and small batch production. Forming processes necessitate component specific and expensive dies since their preparation and design are time-consuming. In latest years, incremental sheet metal forming has acquired good attraction owing to its ability for small-batch productions and prototyping with generic tooling and short lead time. In this paper, a review of literature works has been presented describing investigations carried out on the formability of lightweight alloys, mainly Al alloys, Ti alloys and Mg alloys, during single point incremental forming (SPIF) process, considering the deformation mechanism, formability assessment techniques, forming limit curves, effects of influencing factors, and the effect of heating.

Digital Twin-based Quality Management Method for the Assembly Process of Aerospace Products with the Grey-Markov Model and Apriori Algorithm

The assembly process of aerospace products such as satellites and rockets has the characteristics of single- or small-batch production, a long development period, high reliability, and frequent disturbances. How to predict and avoid quality abnormalities, quickly locate their causes, and improve product assembly quality and efficiency are urgent engineering issues. As the core technology to realize the integration of virtual and physical space, digital twin (DT) technology can make full use of the low cost, high efficiency, and predictable advantages of digital space to provide a feasible solution to such problems. Hence, a quality management method for the assembly process of aerospace products based on DT is proposed. Given that traditional quality control methods for the assembly process of aerospace products are mostly post-inspection, the Grey-Markov model and T-K control chart are used with a small sample of assembly quality data to predict the value of quality data and the status of an assembly system. The Apriori algorithm is applied to mine the strong association rules related to quality data anomalies and uncontrolled assembly systems so as to solve the issue that the causes of abnormal quality are complicated and difficult to trace. The implementation of the proposed approach is described, taking the collected centroid data of an aerospace product’s cabin, one of the key quality data in the assembly process of aerospace products, as an example. A DT-based quality management system for the assembly process of aerospace products is developed, which can effectively improve the efficiency of quality management for the assembly process of aerospace products and reduce quality abnormalities.

Brain–Computer Interface and Hand-Guiding Control in a Human–Robot Collaborative Assembly Task

Collaborative robots (Cobots) are compact machines programmable for a wide variety of tasks and able to ease operators’ working conditions. They can be therefore adopted in small and medium enterprises, characterized by small production batches and a multitude of different and complex tasks. To develop an actual collaborative application, a suitable task design and a suitable interaction strategy between human and cobot are required. The achievement of an effective and efficient communication strategy between human and cobot is one of the milestones of collaborative approaches, which can be based on several communication technologies, possibly in a multimodal way. In this work, we focus on a cooperative assembly task. A brain–computer interface (BCI) is exploited to supply commands to the cobot, to allow the operator the possibility to switch, with the desired timing, between independent and cooperative modality of assistance. The two kinds of control can be activated based on the brain commands gathered when the operator looks at two blinking screens corresponding to different commands, so that the operator does not need to have his hands free to give command messages to the cobot, and the assembly process can be sped up. The feasibility of the proposed approach is validated by developing and testing the interaction in an assembly application. Cycle times for the same assembling task, carried out with and without the cobot support, are compared in terms of average times, variability and learning trends. The usability and effectiveness of the proposed interaction strategy are therefore evaluated, to assess the advantages of the proposed solution in an actual industrial environment.

Single Point Incremental Forming as a Cost-Effective Sheet Forming Process for Small Batch Production

Recently, developing flexible and economical sheet forming processes and technologies is very important. Incremental forming is one of the promising sheet forming processes and it can be valid for these purposes. Especially, single-point incremental forming (SPIF) becomes popular because of its low expenses and rapidly adaption to customers’ needs. In this study, first a brief overview of SPIF will be given and then, outcome of implementing this sheet forming process with minimum costs will be presented. During the study, tool path for forming process is generated using free student version of CAD/CAM software package and low-cost CNC machine used to forming sheet workpiece. Aluminum AA3003 is used as material and combination of oil with grease used as a lubrication for experiments. The result indicated that SPIF is cost-effective process for small batch and prototype production.

Barriers in Implementing Six Sigma in Small and Medium Manufacturing Enterprises

Abstract: Six sigma is a tool that used by organizations to improve quality, Enhance their operation, Improve efficiency as result it helps to increase overall profit, Six sigma remove defects less than 3.4 defects per million, many large companies implement six sigma due to their financial resources ability, that’s why a perception is observed that six sigma is only for large companies, but it can be implemented in small and medium enterprises also with better results. Six sigma is not only for quality but it also helps in increasing productivity, profit, improves whole bottom line of company, this study is for small and medium enterprises that helps to guide for implementation process. Hence Six Sigma is rule which analyzed data and statistics so that industry can focus on primary cause of problem and work accordingly to solve that problem in a simplified manner so that we can get final results improved and better, six Sigma is used in many sectors but gives better result in manufacturing sector also, Small and Medium Enterprises (SMEs) in India operate under typical conditions where a wide variety of products are produced in smaller quantities. For such small batch productions, the production processes including people, machines, materials and methods should be as stable as possible. The quality of the products produced or the services provided must be continuously improved. Whenever possible, inspection should be done at source to avoid quality deviations that would be more costly if found at a later stage.

Implementation of a New Electronic Liquid Dispensing System for Individualized Compounding of Hard Capsules.

An automated compounding device can be a useful tool for the rapid and accurate production of small batches for personalized medicine as well as for clinical batches. A novel electronic liquid dispensing system (ELDS) was investigated to produce hard capsules with individualized dose strengths. An ethanol-based solvent system containing the antihypertensive enalapril maleate was extruded through a cannula into prefilled capsules. The capsules were prefilled with a powder bed of mesoporous silica (Syloid 244 FP) or synthetic dibasic calcium phosphate anhydrous (Fujicalin). The dosing accuracy as well as content uniformity of ELDS was compared with manual preparation using a Hamilton syringe (HS). Both methods met the pharmacopeia criteria for all formulations with an acceptance value (AV) less than 15. Drug adherence to the capsule shells was also investigated. A recovery rate of 98% of enalapril maleate showed almost no drug loss, but the appropriateness of the new dispensing method.

Significant Attention in Industry and Academia for Wire Arc Additive Manufacturing (WAAM) - A Review

The review paper aims to outline the most significant accomplishments in Wire Arc Additive Manufacturing. It can capture the benefits of using the Additive Manufacturing method to manufacture layer-by-layer deposition at significantly reduced production values. Due to the viability of cautiously producing bulky metal components with comparatively high deposition rates, weighty progress has been complete in the thoughtful of the Wire Arc Additive Manufacturing process. Compared to ancient production, the use of this method has resulted in decreased Buy-To-Fly ratios of components the many applications such as saltwater valves, pumps, and turbine propellers, thereby aiding small-batch production. The input parameter is the travel speed, and the output deposition is layered large-scale metal elements with comparatively high deposition rates. The creation of cost-efficient prototypes as microstructure and mechanical properties, deformation, porosity, and cracking are also concerted in different alloy components.

A shuffled cellular evolutionary grey wolf optimizer for flexible job shop scheduling problem with tree-structure job precedence constraints

Along with the growing demands for customized products and small batch production, the flexible job shop manufacturing environment becomes increasingly popular. Efficient flexible job shop scheduling plays a crucial role in making quick responses to production orders with low volume and high variety. When producing complex assembly products that are comprised of multiple and multilevel intermediate parts organized as tree-structure Bills-Of-Materials (BOMs), jobs get restricted by hierarchical precedence constraints due to dependencies between manufactured parts. To cope with this condition, this paper formulates a flexible job shop scheduling problem with job precedence constraints (FJSSP-JPC). A novel shuffled cellular evolutionary grey wolf optimizer (SCEGWO) is proposed to solve FJSSP-JPC with the objective of minimizing makespan. Schedule solutions are encoded as elaborately designed triple-vectors involving the information of job sequencing, grouped operation sequencing and machine assignment, while the satisfactions of job precedence constraints are guaranteed by binary sort tree-based repair mechanism. In SCEGWO, each individual interacts with its topological cellular neighborhood by conducting a micro discrete variant of grey wolf optimizer (GWO), causing that the whole population is decomposed into multiple subpopulations which communicate by the neighborhood overlapping. Extensive experimental results demonstrate that the components of SCEGWO are effective and the proposed SCEGWO outperforms other competing algorithms significantly on the addressed problem.

Current and Future Trends for Polymer Micro/Nanoprocessing in Industrial Applications.

Polymers are widely used in optical devices, electronic devices, energy-harvesting/storage devices, and sensors, owing to their low weight, excellent flexibility, and simple fabrication process. With advancements in micro/nanoprocessing techniques and more demanding application requirements, it is becoming necessary to realize high-resolution fabrication of polymers to prepare miniaturized devices. This is particularly because conventional processing technologies suffer from high thermal stress and strong adhesion/friction, which can irreversibly damage the micro/nanostructures of miniaturized devices. In addition, although the use of advanced fabrication methods to prepare high-resolution micro/nanostructures is explored, these methods are limited to laboratory research or small-batch production. This review focuses on the micro/nanoprocessing of polymeric materials and devices with high spatial precision and replication accuracy for industrial applications. Specifically, the current state-of-the-art techniques and future trends for micro/nanomolding, high-energy beam processing, and micro/nanomachining are discussed. Moreover, an overview of the fabrication and applications of various polymer-based elements and devices such as microlenses, biosensors, and transistors is provided. These techniques are expected to be widely applied for multiscale and multimaterial processing as well as for multifunction integration in next-generation integrated devices, such as photoelectric, smart, and biodegradable devices.

Optimal Design of Flexible Job Shop Scheduling Under Resource Preemption Based on Deep Reinforcement Learning

With the popularization of multi-variety and small-batch production patterns, the flexible job shop scheduling problem (FJSSP) has been widely studied. The sharing of processing resources by multiple machines frequently occurs due to space constraints in a flexible shop, which results in resource preemption for processing workpieces. Resource preemption complicates the constraints of scheduling problems that are otherwise difficult to solve. In this paper, the flexible job shop scheduling problem under the process resource preemption scenario is modeled, and a two-layer rule scheduling algorithm based on deep reinforcement learning is proposed to achieve the goal of minimum scheduling time. The simulation experiments compare our scheduling algorithm with two traditional metaheuristic optimization algorithms among different processing resource distribution scenarios in static scheduling environment. The results suggest that the two-layer rule scheduling algorithm based on deep reinforcement learning is more effective than the meta-heuristic algorithm in the application of processing resource preemption scenarios. Ablation experiments, generalization, and dynamic experiments are performed to demonstrate the excellent performance of our method for FJSSP under resource preemption.